![Sample our Area Studies journals, sign in here to start your access, latest two full volumes FREE to you for 14 days]({"type":"image" , "src" : "/sda/5922/TOC-Subject-Sample-2021-Area-Studies.jpg"})
Abstract
Over the past half century, the Australian public has remained divided on the issue of whether Australia should retain the monarchy or become a republic. Clive Bean found that there had been remarkable stability on the issue and evidence of a long-term trend away from support for the monarchy with a sudden decline in 1992. This article adopts Bean's longitudinal cross-sectional methods to examine the social and political basis of public attitudes. This article analyses the Australian Election Study (1993–2013) to compare Bean's results and re-analyse earlier data from the National Social Science Surveys and Australian National Political Attitudes surveys (1967–90). Public opinion has been fluid and is now at a crossroads between the 1980s high and the 1990s lows. Cohort analysis suggests socialisation impacts long-term opinions. Gender and ethnic nationalism also influences opinion.
过去半个多世纪，澳大利亚公众对于保留君主制还是实行共和制一分为二。克里夫·比恩发现在这个话题上曾经有过明显的稳定性，一直存在脱离君主制的长期趋势，但1992年以来突然式微。本文采用了比恩的长程断面方法来研究公众态度的社会政治基础。本文研究了“澳大利亚选举研究”（1993—2013），同比恩的论点做了比较，并重新分析了“国家社会科学调查”以及“澳大利亚全国政治态度调查”（1967—1990）的数据。公共舆论流变不居，如今正在1980年代高峰与1990年代低谷的交叉点上。同期人群分析显示，社会化影响了长期舆论。性别及族裔民族主义也有其影响。
Acknowledgements
The author is indebted to many who he thanks but especially Rodney Smith, Steven McEachern, Edith Gray, David Smith, Clive Bean, Charles Pattie, Adele Lausberg, Shaun Ratcliff and Marian Simms for their comments, patience and assistance through the article's different stages of development. A previous version of this article was presented at the 2015 Political Studies Association of the United Kingdom annual meeting.
Disclosure statement
No potential conflict of interest was reported by the author.
Notes on contributor
Luke Mansillo is a PhD Candidate in the Department of Government and International Relations at the University of Sydney. He has come to the University of Sydney from the Australian National University, Canberra, where he completed a BA (Hons.) in political science and Masters of Social Research at the School of Politics and International Relations and Australian Demographic and Social Research Institute (ADSRI) respectively. He is the 2015–2016 Australian Political Studies Association Postgraduate Representative.
ORCID
Luke Mansillo http://orcid.org/0000-0002-4366-3993
Notes
1 It is important to note that responses vary to a small degree on the exact question asked. Bean (Citation1993: 192) noted that ANPA, NSSS and AES all have comparable questions which have only ‘minor differences'.
2 The first question has minor differences with the number of response categories. Second there are minor wording changes across the NSSS to the AES and ANPA questions. The author agrees with Bean's opinion (Citation1993: 192) ‘[t]here is no reason to assume that [ … ] these small variations would have any bearing on the comparisons'. The ANPA and AES have three response categories: ‘very important’, ‘fairly important’ and ‘not very important'. The NSSS has the additional category: ‘not important at all’. The 1979 ANPA the question ‘should Australia retain the Queen’ had into two categories. In 1967 only the importance question was asked.
3 Curiously, before 1993, there were significantly more people who wished to retain the monarchy but did not think the Queen was important. This changed by 1993 and probably due to the 1992 royal scandals.
4 Billig (Citation2002: 2–15) wrote of the popular press preoccupation and nothingness of academic press on the issue because sociologists and social psychologists failed to identity monarchy itself as a ‘problem'.
5 Across the 13 surveys with combined measures, the correlations of the two variables are between .7 and .8 and Cronbach's alpha reliability coefficient ranges between .75 and .9. The variables combined make a highly reliable scale.
6 The variables are coded as Bean (Citation1993: 197–98) described his variables.
7 Due to generational replacement the reference category in the reproduction of Bean's cohort analysis was altered.
8 Following the 1992 annus horribilis, the effect for partisanship is comparatively smaller.
9 The finding is reflected in Jackman's (Citation1998: 174) analysis where measures of ethnocentric racial attitudes strongly correlated with support for monarchy.
10 In 1999, a recorded low of 25 per cent of Australians were monarchists if the Australian Constitutional Referendum Study is used, however, respondents were primed by the referendum (McAllister Citation2002: 256).
11 One occasion typifies this sort of activity: in November 1994 Keating detailed royal visit expenses with the aid of a Dorothy Dixer (Hansard Citation1994).
12 The last data point in and is a Newspoll from June 2014, which was 11 weeks after Tony Abbott reintroduced Order of Australia Knights and Dames. The ‘dignified’ part of the state made inroads in opinion, but the ‘efficient’ part could have marred opinion.
13 The Union Flag affixed in Australian flag's corner would impact collective imagining of Australia.
14 Principal component analysis with Varimax rotation used. First three eigenvalues for 1996, 2001 and 2013, respectively, are 2.45, 1.19 and .85; 2.74, 1.26 and .77 and 2.89, 1.21, and .79.
15 Percentages do not always sum exactly to subtotals because of rounding.
16 Dependent variable scaled from 0 (most pro-republic) to 1 (most pro-monarchy).
17 + significant at p≤.10; *p≤.05; ** p≤.01; ***p≤.001. All figures have been rounded to two decimal places, with the exception of age and years of education which are to three decimal places.
18 The means and standard deviations of the dependent variables at each time point show consistency. There are minor measurement differences between the ANPA, NSSS and AES, meaning that the means are not a good guide of the level of public support for the monarchy.
19 The 1990 Election panel file held by the Australian Data Archive no longer has all independent variables; the country of birth is missing from this analysis unlike Bean (Citation1993).